Researcher | Inspur Group Co.,Ltd | China
Peng Wang is a computer architecture researcher serving as a doctoral researcher at the Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, specializing in GPU rendering optimization, compiler optimization, and microarchitecture-independent performance analysis. He has contributed to multiple high-impact research projects, including the development of RayBench and RenderBench benchmark suites, GPU microarchitecture-independent characteristic profiling tools, and LLVM-based RISC-V vectorization optimizations, demonstrating strong technical leadership in benchmarking, compiler engineering, and heterogeneous computing systems. His research outputs include publications in recognized journals such as Electronics, IEEE Access, and other peer-reviewed venues, along with several patents covering GPU performance optimization, cloud game automation, and feature-analysis methodologies. He has actively supported the scientific community through peer-review service for reputable journals and international conferences, reflecting his growing influence in the fields of computer architecture and intelligent systems. His combined expertise in software-hardware co-design, GPU architecture analysis, and compiler technologies positions him as an emerging leader dedicated to advancing high-performance computing research and innovation.
Profile: ORCID
Featured Publications
1. Wang P., Qu H.L., Latency-Aware and Auto-Migrating Page Tables for ARM NUMA Servers. Electronics, 2025, 14(8), 1685.
2. Wang P., Yu Z.B., LLVM RISC-V RV32X Graphics Extension Support and Characteristics Analysis of Graphics Programs. IEEE Access, 2023, 3291920.
3. Wang P., Yu Z., RenderBench: The CPU Rendering Benchmark Suite Based on Microarchitecture-Independent Characteristics. Electronics, 2023, 12(19), 4153.
4. Wang P., Yu Z., RayBench: An Advanced NVIDIA-Centric GPU Rendering Benchmark Suite for Optimal Performance Analysis. Electronics, 2023, 12(19), 4124.
5. Wang P., Chen Y., Xing M.J., Method for Supporting RISC-V Custom Extension Instructions Based on LLVM. Comput. Syst. Appl., 2021, 31(11), Article 8347.
Peng Wang’s work advances high-performance computing by delivering optimized GPU and CPU rendering benchmark suites, compiler enhancements, and microarchitecture-independent performance tools that strengthen the reliability and efficiency of modern computing systems.He aims to drive global innovation through scalable, energy-efficient, and architecture-aware computing solutions that empower future heterogeneous computing technologies.